# ### ## # Parallel Programing In JavaEasy with approach MapReduce

Ali Lotfi edited this page an hour ago · 1 revision Welcome to the ParallelProgramingInJavaEasy- wiki!

Hello. I had a task to do. I had to call a service with different data many times. Each of these calls were independent. Thus, by inspiration from the Hadoop and the MapReduce algorithm, each CPU core As a node, we have a map file that runs on any CPU core, so each map will eventually be merged with a Reduce class.

# ### ## # create map class

public class RestServiceCallMap implements MapJob<Object, Object> {

`public void map(Object input, Context context) throws Exception {`

    `System.out.print(Thread.currentThread().getName());`

    `if (input != null) {`

        `ExcelReadDto shebaNumber = (ExcelReadDto) input;`

        `context.Write(callRestService(shebaNumber.getSheba()));`
}
}
 private ShebaDto callRestService(String sheba) throws Exception {
  
 }

## # create reduce class

public class RestServiceCallReduce implements ReduceJob<Object, Object> {

public void reduce(Object input, Context context) {

    List<ContextImpl> result = (List<ContextImpl>) input;

    List<ShebaDto> flatResult = new ArrayList<ShebaDto>();

    for (ContextImpl item : result) {

        flatResult.addAll((List<ShebaDto>)item.getRepository());

    }

    context.Write(flatResult);

}


} 

## # You can specify the input of the map algorithm

public class ManageInputJobsFile {

public static void addInputPath(final Jobs job, final String pathFile) {
    job.setInputExecuteJob(new InputExecuteJob() {

        public List submit() throws Exception {
            ParserManager excelManager = ExcelManagerFactory.getParserManager(ExcelReadDtosList.class, new ExcelParser(pathFile));
            ResultModel message = excelManager.deSerialize();
            List items = new ArrayList();
            ExcelReadDtosList messageExcel = message.getResultModel();
            for (ExcelReadDto  item : messageExcel.getMessage())
                items.add(item);
            return items;
        }


    });
}
}

## # You can specify the input of the reduce algorithm

public class ManageOutputJobsFile {
public static void addOutputPath(Jobs job, final String pathFile) {
    job.setOutputExecuteJob(new OutPutExecuteJob() {
        public void submit(Object items) throws Exception {

            ExcelReadDtosList excelReadDtosList = new ExcelReadDtosList();
            ArrayList<ShebaDto> items1 = (ArrayList<ShebaDto>) items;
            List<ShebaDto> result = (List<ShebaDto>) items1.get(0);
            for (ShebaDto sb : result) {
                if (sb != null)
                    excelReadDtosList.addMessage(sb.getRequestedIban(), sb.getPaymentCode(), sb.toString());
                else
                    excelReadDtosList.addMessage("null", "error", "eeroor");
            }

            ParserManager excelManagerSerialize = ExcelManagerFactory.getExcelParserManager(pathFile, ExcelReadDtosList.class);
            excelManagerSerialize.serialize(excelReadDtosList);
        }
    });

### ****YOU must create a job class to execute your map/Reduce code

public class RunRestServiceJob implements RunAbleJob {

private Jobs job;


public void run(Object[] args) throws Exception {

    ConfigurationSystem config = new ConfigurationSystem();

    job = new Jobs();

    job.setConfiguration(config);
// job.submit(null);

    job.setMapJob(new RestServiceCallMap());

    job.setReduceJob(new RestServiceCallReduce());

    ManageInputJobsFile.addInputPath(job, "d:\\sheeba.xlsx");

    ManageOutputJobsFile.addOutputPath(job, "d:\\Book1201.xlsx");

    job.waitForCompletion();

}


public core.jobs.Jobs getJobs() {

    return job;

}
}